Elucidating the transcriptional control of osteoblast differentiation is a prerequisite to achieve a better understanding of vertebrate skeletal development in the craniofacial region. Major progress has been made in the last ten years in identifying transcription factors acting as determinant which between mesenchymal cells and osteoblast progenitors in contrast we still have little knowledge how transcription factors control osteoblast terminal differentiation and function. To address this problem we have used a combination of molecular, human and mouse genetic approaches to identify ATF4 as a regulator of osteoblast terminal differentiation and function and the kinase of osteoblast terminal differentiation and function and the kinase RSK2 as a regulator of ATF4 transactivating function. That RSK2 is inactivated in a human skeletal dysplasia termed Coffin Lowry Syndrome that effects primarily the craniofacial region illustrates the biological importance of this regulatory loop. Because ATF4 regulates expression of osteoblast-specific genes and Type I collagen production post-transcriptionally it raises questions about how it executes all its function. Moreover, that RSK2 kinase activity is regulated by extracellular signals suggests that this regulatory loop may explain some of the bone anabolic effect of secreted molecules. Finally, having in hand three osteoblast-specific transcription factors allows us to ask questions about their relationship during development and after birth. To address these questions the specific aims are: To determine to which extent the skeletal phenotype of Atf4-deficient mice is secondary to a decrease in amino acid import. To analyze whether ATF4 is a transcriptional mediator of known extracellular regulators of bone formation. To compare the effect of an osteoblast-specific versus a ubiquitous over-expression of Atf4 in mice. To determine whether Runx2, Osterix and ATF4 interact genetically to affect osteoblast differentiation. To determine whether Atf4 can rescue the osteoblast differentiation defect of Runx2- or of Osterix-deficient mice. To establish in vivo that ATF4 is the physiological target of RSK2 in osteoblasts.